OBJECTIVES: Assess prevalence and discontinuation rates of antidyslipidemic therapy in patients with dyslipidemia. METHODS: We conducted a retrospective cohort study using a large US national database of electronic medical records containing diagnosis, lab and medication information. Users of antidyslipidemics were identified. Prevalence of six classes of antidyslipidemics was described and one-year discontinuation rate was calculated. Logistic regression model was used to evaluate factors affecting drug discontinuation. RESULTS: We identified 153,505 patients who started antidyslipidemic therapy from 1997 to 2004 and were followed for one year after drug initiation. Among 214,440 distinct treatment regimen initiations, 79.5% were statins, with atorvastatin (52.7%) and simvastatin (24.0%) being the most frequently prescribed. Prevalence of other classes of antidyslipidemics was: 8.8% for fibrates, 5.3% for cholesterol absorption inhibitors, 3.3% for niacin, 2.7% for bile acid-binding resins, and 0.5% for combination formulations. One-year discontinuation rate was 28.6% for statins, 31.8% for cholesterol absorption inhibitors, 34.1% for fibrates, 40.7% for combination formulations, 44.7% for niacin, and 50.1% for bile acid-biding resins. Logistic regression showed that statin users were less likely to stop therapy than patients on other classes (odds ratio (OR): 0.63). Surprisingly, patients with coronary hear disease (OR: 1.22) and diabetes (OR: 1.13) were more likely to discontinue treatment than patients without those conditions; high LDL-C or TG and low HDL-C were also found to be associated with higher rates of medication discontinuation at one year. CONCLUSIONS: Our study showed that significant variation exists in medication discontinuation among different classes of antidyslipidemics. Medications targeting to raise HDL-C or lower TG had higher discontinuation rate, implying that there were unmet needs for the treatment of abnormal HDL-C or TG. High discontinuation rate for patients at high risk for cardiovascular disease was a major concern. Further studies are needed to assess the consequences and implications of medication discontinuation.